Geostatistical Modeling and Heterogeneity Analysis of Tumor Molecular Landscape

SIMPLE SUMMARY: The present study introduces a new computational platform referred to as GATHER to conduct Geostatistical Analysis of Tumor Heterogeneity and Entropy in R. GATHER has several distinct advantages such as (a) a novel use of single-cell-specific spatial information for kriging to synthesize high-resolution and continuous gene expression landscapes of a given tumor sample, (b) the integration of such landscapes to identify and map the enriched regions of the pathways of interest, (c) the identification of genes that have a spatial differential expression at locations representing specific phenotypic contexts, (d) the computation of spatial entropy measures for quantification and objective characterization of intratumor heterogeneity, and (e) the use of new tools for the insightful visualization of spatial transcriptomic phenomena. ABSTRACT: Intratumor heterogeneity (ITH) is associated with therapeutic resistance and poor prognosis in cancer patients, and attributed to genetic, epigenetic, and microenvironmental factors. We developed a new computational platform, GATHER, for geostatistical modeling of single cell RNA-seq data to synthesize high-resolution and continuous gene expression landscapes of a given tumor sample. Such landscapes allow GATHER to map the enriched regions of pathways of interest in the tumor space and identify genes that have spatial differential expressions at locations representing specific phenotypic contexts using measures based on optimal transport. GATHER provides new applications of spatial entropy measures for quantification and objective characterization of ITH. It includes new tools for insightful visualization of spatial transcriptomic phenomena. We illustrate the capabilities of GATHER using real data from breast cancer tumor to study hallmarks of cancer in the phenotypic contexts defined by cancer associated fibroblasts.